app.py

import gradio as gr
import requests
import os

# Define the Hugging Face API endpoint
api_url = "https://api-inference.huggingface.co/models/Qwen/Qwen2.5-72B-Instruct"

# Retrieve the token from the environment
headers = {"Authorization": f"Bearer {os.getenv('HF_Token')}"}

# Define a function to send prompts to the model and get responses
def query(prompt):
    response = requests.post(api_url, headers=headers, json={"inputs": prompt})
    if response.status_code == 200:
        return response.json()[0]["generated_text"]
    else:
        return f"Error {response.status_code}: {response.text}"

# Create a Gradio interface with text input and output
demo = gr.Interface(
    fn=query,
    inputs="text",
    outputs="text",
    title="Qwen-2.5 72B Interaction",
    description="Ask complex mathematical or pattern-related questions and get responses from Qwen-2.5 72B."
)

# Launch the interface
demo.launch()